Original Article

Gender determination by odontometric

method
Litha, H. C. Girish1,
Sanjay Murgod1,
J. K. Savita1
Department of Oral Pathology Abstract
and Microbiology, Farooqia
Dental College and Hospital, Context: Gender determination is central in establishing personal identification from
Mysore, 1Department of Oral human skeletal remains. The study was conducted to find out the accuracy with
Pathology and Microbiology, which gender can be determined by odontometric methods. Aims: To investigate the
RajaRajeswari Dental College and
Hospital, Bengaluru, mesiodistal (MD) and buccolingual (BL) dimensions of all the teeth of permanent dentition
Karnataka, India to find new parameters to differentiate between male and female teeth and to assess
whether each type of linear measurement can be used independently in odontometric
sex differentiation. Materials and Methods: The study was conducted at a dental
college on a composite group of 500 individuals comprising 250 males and 250 females.
Impressions of upper and lower jaws were made with alginate impression material and
casts prepared with dental stone. A digital Vernier calliper was used to measure the BL
and MD dimensions of all the upper teeth except the third molars. Statistical Analysis
Used: The results were subjected to statistical analysis using univariate analysis and
linear stepwise discriminant function analysis to find the variables which discriminate
gender significantly. Results: The MD and BL dimensions between males and females
were statistically significant. The predicted value for correct classification of gender
Address for correspondence: was also statistically significant. Conclusions: The ability to differentiate gender in the
Dr. Litha, population using stepwise discriminant functions was found to be very high with 99.8%
Department of Oral Pathology
accuracy with males showing statistically larger teeth than females. This is similar to the
and Microbiology, Farooqia
Dental College and Hospital, near 100% success in gender determination using pelvic and skull bones.
Mysore, Karnataka, India.
E‑mail: litha4excellence@gmail. Key words: Buccolingual, gender determination, mesiodistal, odontometric
com

Introduction Technically, the positive identification of human remains

requires the matching of physical characteristics of the

T he identification of individuals and their unique deceased with records of the physical characteristics made
characteristics have been of paramount importance to before death. Several methods may be used to produce
human society. It is important to identify the deceased to identification with an aggregate high level of reliability.
ensure a proper burial and for the satisfactory disposal of Identification by relatives and friends may not be accurate
legal issues that might arise.[1] because of the highly emotional state of the relatives and

Access this article online This is an open access article distributed under the terms of the Creative
Quick Response Code Commons Attribution‑NonCommercial‑ShareAlike 3.0 License, which allows
others to remix, tweak, and build upon the work non‑commercially, as long as the
Website:
author is credited and the new creations are licensed under the identical terms.
www.jfds.org
For reprints contact: reprints@medknow.com

DOI: How to cite this article: Litha, Girish HC, Murgod S, Savita JK.
10.4103/jfo.jfds_96_15 Gender determination by odontometric method. J Forensic Dent Sci
2017;9:44-5.
 Litha, et al.: Gender determination by odontometric method

also because sometimes, the bodies may be highly mutilated environmental influences.[6] Therefore, this study was
or decomposed.[1] undertaken to assess the dimorphism of permanent teeth
in humans by measuring the BL and MD diameter of the
The most reliable method of identification includes finger teeth. The results obtained will be subjected to stepwise
prints and biological methods such as DNA profiling. In discriminant functional analysis to enable accurate sex
some cases, fingerprints may not be available due to trauma assessment in forensic identification.
and decomposition. DNA profiling exhibits a high degree
of dependability, but can be expensive, time‑consuming, Materials and Methods
and may not be feasible in all cases.
This clinical study was conducted over a period of 4 months
Dental examination and comparison between antemortem on undergraduate and postgraduate students of a dental
and postmortem dental records and radiographs produce college. The sample comprised permanent dentitions from
results with a high degree of reliability and relative 500 individuals of Indian origin, of which 250 were males
simplicity. Teeth are the most durable structures in the and 250 females. The sample was the composite of different
body and can resist temperatures of 1600°C without any ethnic groups as participants were from different states of
appreciable loss of microstructure. Hence, teeth form an India.
excellent material for anthropological, genetic, odontologic,
and forensic investigations.[1,2] Inclusion criteria
1. Healthy state of gingiva and periodontium
The term “forensic” implies “court of law,” “odontology” 2. Caries‑free teeth.
means “study of tooth.” Forensic odontology has been
defined as the branch of dentistry which, in the interest of Exclusion criteria
justice, deals with the proper handling and examination 1. Supernumerary teeth
of dental evidence and with the proper evaluation and 2. Third molars
presentation of dental findings. Forensic odontology has 3. Developmental abnormalities of teeth
played a key role in the identification of persons in mass 4. Physical and chemical injuries of teeth
disasters, in crime investigations, and in investigations of 5. Teeth with proximal restorations.
decomposed and disfigured bodies.[3]
Materials required
Various methods are employed in forensic odontology to a. For examination and selection of study subjects, the
determine the age and sex of an individual. The methods following materials are required:
include rugoscopy, cheiloscopy, bite marks, tooth prints, 1. Disposable mouth mask
radiographs, photographic study, and molecular methods.[3] 2. Disposable gloves
3. Chlorhexidine mouthwash
The determination of gender is central to the process of 4. Cheek retractor
establishing personal identification from human skeletal 5. Stainless steel kidney tray
remains. An accurate sex diagnosis effectively reduces 6. Cotton holder and cotton
the number of possible matches by half. It also helps in 7. Disposable glass tumbler
estimating the stature and age at the death of an individual. 8. Tweezer
Measurement of the long bones may also provide highly 9. Mouth mirror
accurate sex assessments, but may not be possible in all 10. Single‑ended straight sharp probe
cases.[4] 11. Pro forma and consent form.

In some of the cases, the only available structure for b. For impression making and obtaining study cast, the
determining gender is the measurement of the permanent following materials are required:
dentition. Acharya and Mainali explored the utility of 1. Dentulous perforated stock trays for full mouth
buccolingual (BL) dimension and mesiodistal (MD) impressions – upper and lower number 1, 2, 3, 4
dimension in sex differentiation and reported accuracy rates 2. Rubber bowls
of 62–83% for a Nepalese sample.[5] Ates et al. observed an 3. Plaster spatulas – straight and curved
accuracy of 68–81% for similar measurements in a Turkish 4. Alginate impression material – Algitex
sample. 5. Type III Gypsum product, dental stone – Denstone
6. Type II Gypsum product, dental plaster – Dentaplast.
Although teeth are excellent study material, variation in the
crown size has been reported among different populations. c. For recording the measurements on the obtained study
Numerous factors can contribute to variation in tooth size cast, the following material is required:
and may be described broadly as genetic, epigenetic, and 1. Digital Vernier calliper calibrated to 0.01 mm.
 Litha, et al.: Gender determination by odontometric method

Method The gender distribution of cases is represented in Table 1.

After obtaining written consent from the participants, Out of 500 cases, 250 (50%) cases were males and 250 (50%)
impressions of the maxillary and mandibular arches were cases were females.
taken using dental alginate. Casts were poured using dental
stone. The descriptive statistics, t‑values, and P values of MD
measurements of the permanent teeth in males and females
Measurements are depicted in Table 2. All 28 teeth variables were included
The MD [Figure 1] and BL [Figure 2] dimensions of all the in the analysis. Results on continuous measurements are
teeth except third molars were measured on the casts using presented on mean ± standard deviation (SD) (Min‑Max)
a digital Vernier calliper calibrated to 0.01 mm (Mitsoyo, and results on categorical measurements are presented
Japan).[6,7] The measurements were noted down on the master in number (%). Significance is assessed at 5% level of
chart and later entered in Microsoft excel spreadsheet and significance. Statistically strongly significant values,
were subjected to statistical analysis. The MD measurement with P ≤ 0.01 was seen in MD values of 13 (P = 0.001),
was defined as the greatest distance between contact points 21 (P = 0.001), 36 (P = 0.003), and 33 (P = 0.004). Statistically
on the approximate surfaces of the tooth crown and was moderately significant values, with P > 0.01 and P ≤ 0.05 are
measured with the caliper beaks perpendicular to the long noted in the MD values of 43 (P = 0.024) and 46 (P = 0.017),
axis of the tooth. The BL measurement was defined as the respectively. MD value of 45 with P = 0.088 had statistically
greatest distance between the labial/buccal surfaces of the suggestive significance.
tooth crown measured with the caliper beaks held at the
right angles to the MD dimensions.[7] Figure 3 shows the MD odontometrical variables of all teeth
in males and females. MD dimensions of male teeth are
Results greater than the MD dimensions of female teeth in 17, 16,
13, 12, 11, 21, 22, 23, 24, 26, 27, 36, 35, 34, 33, 43, 45, and 46.
An observational multivariate discriminate study with
250 males and 250 females was undertaken to determine The descriptive statistics, t‑values, and P values of BL
the significant odontometric variables for discriminating measurements of the permanent teeth in males and
gender.

Statistical analysis Table 1: Gender distribution

Linear stepwise discriminant function analysis has been Gender Number of subjects (%)
performed to find the variables which discriminate the Mesiodistal
gender significantly. All statistical analyses were performed Male 250 (50.0)
using statistical software, namely, SAS 9.2 9 (SAS Institute Inc.), Female 250 (50.0)
SPSS 15.0 (IBM Corp), Stata 10.1 (Oracle), MedCalc 9.0.1 Total 500 (100.0)
(Med Calc Software), Systat 12.0 (Systat Software), and R Buccolingual
Environment version 12.11.1 (R Foundation for Statistical Male 250 (50.0)
Computing). Microsoft Word and Excel have been used to Female 250 (50.0)
generate tables and graphs. Total 500 (100.0)

Figure 1: Measurement of mesiodistal width of the study cast Figure 2: Measurement of buccolingual width of the study cast
 Litha, et al.: Gender determination by odontometric method

females are depicted in Table 3. All 28 teeth variables Table 4 depicts the tooth variables that contributed to the
were included in the analysis. Results on continuous stepwise discriminant analysis for MD dimensions. Wilks’
measurements are presented on mean ± SD (Min‑Max) lambda denotes how useful a given variable is in the
and results on categorical measurements are presented stepwise analysis and determines the order in which the
in number (%). Significance is assessed at 5% level of variables enter the analysis. MD dimension of maxillary
significance. Statistically strongly significant values, right canine entered the discriminant analysis first followed
with P ≤ 0.01 was seen in BL values of 16 (P = 0.001), 15
(P = 0.002), 14 (P = 0.001), 13 (P = 0.005), 12 (P < 0.001), 11
(P = 0.001), 25 (P < 0.001), 26 (P < 0.001), 27 (P = 0.001),
36 (P = 0.003), 41 (P < 0.001), and 46 (P < 0.001). Statistically
moderately significant values, with P > 0.01 and P ≤ 0.05
is noted in the BL value of 17 (P = 0.026). BL values of 21
(P = 0.096) and 37 (P = 0.055) had statistically suggestive
significance.

Figure 4 shows the BL odontometrical variables of all teeth

in males and females. BL dimensions of male teeth are
greater than the MD dimensions of female teeth in 17, 16,
15, 14, 13, 12, 11, 21, 22, 24, 25, 26, 27, 37, 36, 35, 34, 33, 32, Figure 3: Representation of the mesiodistal odontometric variables of
41, 43, 44, 45, 46, and 47. all the 28 teeth in males and females

Table 2: Univariate analysis of odontometrical Table 3: Univariate analysis of odontometrical

variables (mesiodistal) in males and females variables (buccolingual) in males and females
Variables Male Female Significance Variables Male Female Significance
Mean SD CV% Mean SD CV% Mean SD CV% Mean SD CV%
MD17 9.70 0.88 9.04 9.55 0.77 8.09 t=2.002; P=0.046 BL17 10.65 0.79 7.41 10.50 0.67 6.42 t=2.236; P=0.026*
MD16 9.99 0.66 6.64 9.87 0.61 6.16 t=2.131; P=0.034 BL16 10.72 0.71 6.58 10.52 0.65 6.21 t=3.280; P=0.001**
MD15 6.48 0.81 12.49 6.55 0.76 11.63 t=0.986; P=0.324 BL15 9.11 0.81 8.84 8.88 0.82 9.27 t=3.110; P=0.002**
MD14 6.75 0.53 7.83 6.79 0.55 8.03 t=0.801; P=0.423 BL14 8.94 0.66 7.34 8.74 0.70 8.00 t=3.243; P=0.001**
MD13 7.63 0.60 7.85 7.45 0.60 8.10 t=3.325; P=0.001** BL13 7.16 0.91 12.73 6.94 0.86 12.45 t=2.808; P=0.005**
MD12 6.83 0.66 9.64 6.80 0.60 8.88 t=0.357; P=0.721 BL12 6.80 0.60 8.9 5.63 1.03 18.25 t=15.556; P<0.001**
MD11 8.52 0.76 8.96 8.39 0.76 9.11 t=1.899; P=0.058 BL11 6.15 1.01 16.40 5.91 1.09 18.53 t=2.579; P=0.001**
MD21 8.63 0.68 7.85 8.44 0.65 7.74 t=3.249; P=0.001** BL21 6.14 1.06 17.25 5.98 1.05 17.49 t=1.666; P=0.096+
MD22 6.82 0.70 10.21 6.76 0.75 11.15 t=0.967; P=0.334 BL22 5.57 0.98 17.66 5.50 1.08 19.57 t=0.734; P=0.463
MD23 7.54 0.69 9.20 7.44 0.62 8.30 t=1.598; P=0.111 BL23 6.99 0.96 13.75 7.00 1.26 18.00 t=0.052; P=0.958
MD24 6.88 0.62 9.00 6.84 0.59 8.63 t=0.675; P=0.500 BL24 8.84 0.80 9.11 8.63 0.81 9.42 t=2.851; P=0.005**
MD25 6.42 0.74 11.49 6.51 0.74 11.42 t=1.437; P=0.151 BL25 9.09 0.75 8.21 8.78 0.81 9.21 t=4.515; P<0.001**
MD26 9.74 0.91 9.30 9.81 0.67 6.81 t=1.113; P=0.266 BL26 10.73 0.87 8.10 10.41 0.81 7.75 t=4.280; P<0.001**
MD27 9.66 0.88 9.07 9.62 0.79 8.16 t=0.544; P=0.587 BL27 10.70 0.78 7.25 10.48 0.78 7.40 t=3.204; P=0.001**
MD37 9.95 0.73 7.37 9.89 0.76 7.63 t=0.862; P=0.389 BL37 9.90 0.84 8.49 9.77 0.66 6.81 t=1.925; P=0.055+
MD36 10.76 0.86 7.99 10.52 0.94 8.93 t=2.956; P=0.003** BL36 10.21 0.70 6.86 10.02 0.75 7.50 t=3.014; P=0.003**
MD35 6.94 0.79 11.36 6.96 0.70 10.02 t=0.368; P=0.713 BL35 8.09 0.73 9.01 8.03 0.67 8.31 t=0.949; P=0.343
MD34 6.98 0.64 9.22 6.94 0.58 8.31 t=0.599; P=0.550 BL34 7.40 0.73 9.90 7.35 0.69 9.40 t=0.855; P=0.393
MD33 6.69 0.56 8.31 6.54 0.54 8.28 t=2.906; P=0.004** BL33 6.28 0.93 14.82 6.20 0.82 13.25 t=1.013; P=0.311
MD32 5.87 0.57 9.68 5.86 0.56 9.51 t=0.191; P=0.849 BL32 5.38 0.89 16.63 5.30 0.79 14.95 t=1.072; P=0.284
MD31 5.37 0.61 11.40 5.42 0.70 13.00 t=0.823; P=0.411 BL31 5.08 0.91 17.85 5.06 0.89 17.65 t=0.192; P=0.848
MD41 5.34 0.54 10.09 5.41 0.69 12.75 t=1.334; P=0.182 BL41 5.41 0.69 12.75 5.09 0.87 17.11 t=4.760; P<0.001**
MD42 5.82 0.55 9.41 5.88 0.58 9.90 t=1.141; P=0.255 BL42 5.30 0.91 17.24 5.29 0.91 17.14 t=0.133; P=0.895
MD43 6.65 0.67 10.09 6.52 0.63 9.63 t=2.259; P=0.024* BL43 6.23 0.91 14.62 6.16 0.96 15.62 t=0.892; P=0.373
MD44 6.91 0.60 8.71 6.96 0.55 7.91 t=1.017; P=0.309 BL44 7.38 0.74 10.02 7.31 0.77 10.57 t=0.945; P=0.345
MD45 6.86 0.80 11.62 6.98 0.78 11.17 t=1.712; P=0.088+ BL45 8.06 0.72 8.91 7.99 0.76 9.56 t=1.146; P=0.252
MD46 10.67 0.90 8.42 10.47 0.95 9.10 t=2.340; P=0.017* BL46 10.47 0.95 9.01 9.94 0.85 8.58 t=4.316; P<0.001**
MD47 10.04 0.78 7.77 9.99 1.09 10.96 t=0.582; P=0.561 BL47 9.99 1.09 10.90 9.99 0.72 7.21 t=0.0316; P=0.969
MD: Mesiodistal, SD: Standard deviation, CV: Coefficient of variance, **: Strongly BL: Buccolingual, SD: Standard deviation, CV: Coefficient of variance, **: Strongly
significant, *: Moderately significant, +: Suggestive significance significant, *: Moderately significant, +: Suggestive significance.
 Litha, et al.: Gender determination by odontometric method

Figure 4: Representation of the buccolingual odontometrical variables of all the 28 teeth in males and females

Table 4: Stepwise discriminant function analysis (mesiodistal)

Variables Unstandardized coefficients Standardized coefficients Sectioning point Wilks’ lambda P Percentage of
correct classification
MD13 0.775 0.466 0.278 for male −0.276 0.869 <0.001** 69.9
MD21 0.691 0.460 for female
MD36 0.554 0.490
MD42 −0.771 −0.430
MD45 −0.660 −0.501
MD: Mesiodistal, **: Strongly significant

by maxillary left central incisor, mandibular left first molar, stepwise analysis and determines the order in which the
mandibular right lateral incisor, and mandibular right variables enter the analysis. MD dimension of maxillary
second premolar. The unstandardized coefficients are 0.775 left central incisor entered the discriminant analysis first
for MD13, +0.691 for MD21, +0.554 for MD36, −0.771 for followed by MD dimension of mandibular left central
MD42, and − 0.660 for MD45. Sectioning point is 0.278 for incisor, BL dimension of maxillary right second molar,
male and − 0.276 for female. Wilks’ lambda is 0.869 with maxillary right second premolar, maxillary right first
69.9% predicted value for correct classification and this is premolar, maxillary right lateral incisor, maxillary right
statistically significant. central incisor, maxillary left central incisor, maxillary
left first premolar, maxillary left second premolar,
Table 5 depicts the tooth variables that contributed to the mandibular left second molar, mandibular left first molar,
stepwise discriminant analysis for BL dimensions. Wilks’ and mandibular left second premolar. The unstandardized
lambda denotes how useful a given variable is in the coefficients are ‑0.207 for MD 21, +0.209 for MD 31, +0.190
stepwise analysis and determines the order in which the
for BL 17, +0.484 for BL 15, +0.683 for BL 14, −0.168 for BL
variables enter the analysis. BL dimension of maxillary
12, −0.219 for BL 11, −0.510 for BL 21, +0.297 for BL 24, +0.214
right second premolar entered the discriminant analysis
for BL 25, −0.202 for BL 37, −0.323 for BL 36, and +0.243 for
first followed by maxillary right first premolar, maxillary
BL 35. Sectioning point is −3.172 for males and 3.159 for
right canine, maxillary right central incisor, maxillary left
females. Wilks’ lambda is 0.090 with 99.8% predicted value
central incisor, maxillary left lateral incisor, maxillary left
first premolar, maxillary left second premolar, maxillary left for correct classification and this is statistically significant.
second molar, mandibular left second molar, mandibular Hence, this function can predict the new case as male or
left first molar, and mandibular left second premolar. The female with 99.8% accuracy.
unstandardized coefficients are 0.469 for BL 15, +0.731 for
BL 14, −0.143 for BL 13, −0.263 for BL 11, −0.426 for BL 21, Discussion
−0.151 for BL 22, +0.265 for BL 24, +0.223 for BL 25, +0.212
for BL 27, −0.186 for BL 37, −0.312 for BL 36, and + 0.251 Proper sex assessment of skeletal remains has important
for BL 35. Sectioning point is − 3.137 for male and 3.137 for ramifications in forensic and bioarchaeological
female. Wilks’ lambda is 0.092 with 99.8% predicted value investigations. All the available criteria must be utilized
for correct classification and is statistically significant. in order to obtain optimal sex prediction. Teeth can
resist postmortem insults and are considered one of the
Table 6 depicts the tooth variables that contributed to the strongest tissues in the body. Hence, teeth are of paramount
stepwise discriminant analysis for BL and MD dimensions. importance when more robust predictors such as the pelvis
Wilks’ lambda denotes how useful a given variable is in the or long bones are destroyed or fragmented.[7]
 Litha, et al.: Gender determination by odontometric method

Table 5: Stepwise discriminant function analysis (buccolingual)

Variables Unstandardized coefficients Standardized coefficients Sectioning point Wilks’ Lambda P Percentage of
correct classification
BL15 0.469 0.368 −3.137 for male 0.092 <0.001** 99.8
BL14 0.731 0.441 3.137 for female
BL13 −0.143 −0.111
BL11 −0.263 −0.235
BL21 −0.426 −0.375
BL22 −0.151 −0.132
BL24 0.265 0.187
BL25 0.223 0.166
BL27 0.212 0.166
BL37 −0.186 −0.149
BL36 −0.312 −0.259
BL35 0.251 0.179
BL: Buccolingual, **: Strongly significant

Table 6: Stepwise discriminant function analysis (mesiodistal + buccolingual)

Variables Unstandardized coefficients Standardized coefficients Sectioning point Wilks’ lambda P Percentage of
correct classification
MD21 −0.207 −0.138 −3.172 for male 0.090 <0.001** 99.8
MD31 0.209 0.138 3.159 for female
BL17 0.190 0.148
BL15 0.484 0.378
BL14 0.683 0.412
BL12 −0.168 −0.139
BL11 −0.219 −0.196
BL21 −0.510 −0.449
BL24 0.297 0.210
BL25 0.214 0.160
BL37 −0.202 −0.162
BL36 −0.323 −0.267
BL35 0.243 0.173
MD: Mesiodistal, BL: Buccolingual, **: Strongly significant

Human dentition has been subjected to many studies, The aim of the study was to find out the accuracy with which
and most of the studies have been focused on human gender can be differentiated by investigating the MD and
dental health, morphology, and odontometric variation. BL dimensions of all teeth of the permanent dentition. The
Forensically, teeth have been used mostly for age estimation study was conducted on a composite group of 500 students
and sex determination.[8] Out of the two approaches to comprising 250 males and 250 females.
identify sex, the first is based on a visual assessment of the
shape or relative proportions of sexually dimorphic features. Sexual dimorphism of teeth has been studied by many
The second is a metric approach, which offers advantages researchers and it has been reported that permanent teeth
over the visual approach as it is inherently more objective, are highly dimorphic.[6,11] In our study, both the jaws were
has higher reliability, is less dependent on the previous utilized as it is known to provide the highest accuracy rate
observer experience, and is more readily amendable to in assessing sex.[12] In all the teeth in which the MD and BL
statistical analysis, and thus helps comparisons within the variables showed statistically significant difference, males
samples as well as with previous studies. The MD and BL had larger tooth dimensions when compared to that of
diameters of permanent teeth are the two most commonly females. This finding reinforces the similar conclusions
used and researched features used in the determination drawn by Vanaki et al., Zorba et al., Pereira et al., and Iscan
of sex on the basis of dental measurements.[9] Most of the and Kedici that males exhibit larger teeth than females.[8‑11]
odontometric sex determinations have been based on Zorba et al. have compiled the various reasons put forward
the use of stepwise discriminant function analysis as it by researchers to explain the increased dimensions in male
calculates the optimal combination of variables and weighs teeth. It includes an increase in jaw size of males, difference
them to reflect their contribution to sex determination.[10] in enamel, and the role of Y chromosome in increasing the
 Litha, et al.: Gender determination by odontometric method

mitotic potential of the tooth germ which, in turn, induces On the other hand, the MD dimension of 42 did not
dentinogenesis.[11] show significant univariate difference, yet entered the
discriminant analysis. Similarly, when only the BL variable
Out of the 56 variables, 10 variables exhibited reverse was used, teeth with significant univariate dimorphism
dimorphism, i.e. an increase in the dimensions of female which contributed to the discriminant analysis were 15,
teeth when compared to the dimensions of male teeth, but 14, 13, 11, 21, 25, 27, 36, and 35. Whereas, 17, 16, 12, 21,
it did not affect the sex assessment as none of the variables 26, 35, 41, and 46 did not enter the discriminant functions
were statistically significant and hence did not contribute to although they showed univariate differences. The BL
the stepwise discriminant function. The feature of reverse dimensions of 22, 24, and 27 did not show significant
dimorphism has also been reported in a similar study by univariate difference, but entered the discriminant function.
Prabhu and Acharya. Multifarious interactions between This is reflected in an earlier work by Acharya and Mainali
different genetic and environmental factors and the reduced where a majority of BL dimensions that showed significant
sexual dimorphism and consequent male‑female overlap univariate differences did not enter the discriminant
have probably extended to include larger female teeth and functions.[12] This suggests that univariate (independent
result in reverse dimorphism.[12,13] t‑test) and multivariate (discriminant) analysis may give
different results when used for sexing.
Of all the permanent teeth, many studies have reported that
canines are the most dimorphic teeth.[12,14‑16] The present The BL variables were systematically better in sex identification
study establishes the existence of a statistically significant than MD dimensions. The stepwise discriminant analysis for
difference in the BL measurements of all the teeth in BL variables differentiated the sex with an accuracy of 99.8%
maxillary right quadrant, maxillary left second premolar, while for MD dimensions, the accuracy was 69.9%. The
first molar and second molar, mandibular right and left increased accuracy with which BL variables differentiated sex
first molar, and mandibular right central incisor showed when compared to MD variables is similar to that reported
statistically significant difference in the crown size between by Prabhu and Acharya and Garn et al.[11,13,17] The latter
male and female teeth. Mesiodistally, the maxillary and recommended wider use of BL dimensions. It is plausible that
mandibular canines, mandibular first molar, and maxillary the inability of additional MD variables to enter the stepwise
central incisor showed pronounced sexual dimorphism. discriminant analysis is responsible for their relatively low
This is in accordance with the studies conducted by Acharya accuracy as was noticed in the study conducted by Acharya
and Mainali who reported that sexual dimorphism is and Mainali in which there was a low accuracy of BL variables
more in canines, molars, and maxillary central incisors.[9,12] due to the inability of many BL variables to enter the stepwise
Studies by Garn et al. have also reported a significant sexual discriminant analysis.[5]
dimorphism in canines and molars.[14] The anterior teeth
did not contribute in discriminating the sexes as much as When the MD and BL variables were combined and used,
the posterior teeth. This was in contrast to the observations the posterior teeth discriminated the sexes more than the
of Iscan and Kedici wherein, anterior teeth discriminated anterior ones. Although the MD variables entered the
the sexes in a better way.[8] Variation in sample size can be stepwise discriminant function first, 11 out of 13 variables
considered to be a reason for such a diametrically opposed that entered the analysis were BL dimensions. About 99.8%
observation. Maxillary teeth entered the discriminant accuracy in sexing is seen when the BL and MD variables
analysis more than the mandibular teeth just as in the study are used together. This is the highest among similar
conducted by Acharya and Mainali.[12] studies.[5,7‑13] Increased sample size in our study can be
one reason for this. Another reason can be ethnic mixing
In similar studies, the BL measurements were found to be which causes changes in dental dimensions and hence gives
more dimorphic than the MD measurements.[6,10‑12] Results varying results.[6‑12,18]
of our studies have confirmed this finding. The reason for
this can be because with time, the MD dimension can be Conclusions
influenced by advanced consumption of special foods and
the approximate surfaces could show signs of wear, which, Discriminant analysis that predicted value for correct
in turn, may give altered dental measurements and impact classification is as follows:
sex assessment outcomes.[7] 1. MD variables had 69.9% accuracy in predicting sex with
statistically significant value
When only the MD variable was used, majority of tooth 2. BL variables had 99.8% accuracy in predicting sex with
variables such as 13, 21, 36, and 34, which showed statistically significant value
statistically significant univariate dimorphism have 3. MD and BL variables when combined had 99.8%
contributed to the discriminant analysis. However, the accuracy for predicting sex with statistically significant
MD dimensions of 33, 43, and 46 that showed univariate value similar to that found when only BL variables are
differences did not enter the discriminant functions. used.
 Litha, et al.: Gender determination by odontometric method

The present study revealed the following findings: 2. Rai B, Anand SC. Gender determination by diagonal distances of
1. Significant dimorphic differences between male and teeth. Internet J Biol Anthropol 2007;1.
female teeth with males exhibiting larger teeth than 3. Kavitha B, Einstein A, Sivapathasundharam B, Saraswathi TR.
Limitations in forensic odontology. J Forensic Dent Sci 2009;1:8‑10.
females
2. Posterior teeth discriminated the sexes slightly more 4. Macaluso PJ Jr. Sex discrimination potential of permanent maxillary
molar cusp diameters. J Forensic Odontostomatol 2010;28:22‑31.
than the anterior teeth
5. Acharya AB, Mainali S. Sex discrimination potential of buccolingual
3. Maxillary teeth entered the discriminant analysis more
and mesiodistal tooth dimensions. J Forensic Sci 2008;53:790‑2.
than the mandibular teeth
6. Suazo GL, Cantin LM, Lopez FB, Sandoval MC, Torres MS,
4. BL variables, as a unit, provide greater accuracy in Gajardo RP . Sexual dimorphism in mesiodistal and buccolingual
gender determination when compared to MD variables tooth dimensions in Chilean people. Int J Morphol 2008;26:609‑14.
5. The accuracy in predicting gender by BL measurements 7. Acharya AB, Prabhu S, Muddapur MV. Odontometric sex
alone is same as the accuracy obtained when both the assessment from logistic regression analysis. Int J Legal Med
MD and BL variables are considered together 2011;125:199‑204.
6. Few teeth showed reverse dimorphism, but were not 8. Iscan MY, Kedici PS. Sexual variation in bucco‑lingual dimensions
statistically significant. in Turkish dentition. Forensic Sci Int 2003;137:160‑4.
9. Vanaki SS, Puranik RS, Sharma G, Sharma M. Tooth dimension as a
distinguishing trait between human sexes: An odontometric study
One of the greatest challenges faced by forensic experts is
on Bagalkot population. Indian J Forensic Med Pathol 2008;1:75‑8.
determination of gender using skeletal remains, especially
10. Pereira C, Bernardo M, Pestana D, Santos JC, Mendonça MC.
when only fragments of the body are recovered. Contribution of teeth in human forensic identification – discriminant
function sexing odontometrical techniques in Portuguese
The results obtained in this study support the observations population. J Forensic Leg Med 2010;17:105‑10.
made by a plethora of previous studies that male tooth 11. Zorba E, Moraitis K, Manolis SK. Sexual dimorphism in permanent
dimensions are statistically larger than females. The teeth of modern Greeks. Forensic Sci Int 2011;210:74‑81.
ability to differentiate sex in the population using stepwise 12. Acharya AB, Mainali S. Univariate sex dimorphism in the
discriminant functions was found to be very high with Nepalese dentition and the use of discriminant functions in gender
assessment. Forensic Sci Int 2007;173:47‑56.
99.8% accuracy. This is similar to the near 100% success in
13. Prabhu S, Acharya AB. Odontometric sex assessment in Indians.
sex assessment using pelvic and skull bones, although the
Forensic Sci Int 2009;192:129.e1‑5.
classification accuracy of most functions ranged between
14. Garn SM, Lewis AB, Swindler DR, Kerewsky RS. Genetic control
70% and 90%. Consequently, sexing from tooth dimensions of sexual dimorphism in tooth size. J Dent Res 1967;46:963‑72.
has always been considered as an adjunct rather than
15. Al‑Rifaiy MQ, Abdullah MK, Ashraf I, Khan N. Dimorphism of
the sole indicator of gender. Documentation of similar mandibular and canine establishing sex identity. Saudi Dent J
observations with further studies will enable the use of each 1997;9:1‑5.
type of linear measurement independently in odontometric 16. Boaz K, Gupta C. Dimorphism in human maxillary and mandibular
sex differentiation. canines in establishment of gender. J Forensic Dent Sci 2009;1:42‑4.
17. Garn SM, Lewis AB, Kerewsky RS. Sexual dimorphism in the
References buccolingual tooth diameter. J Dent Res 1966;45:1819.
18. Brook AH, Griffin RC, Townsend G, Levisianos Y, Russell J,
1. Rothwell BR. Principles of dental identification. Dent Clin North Smith RN. Variability and patterning in permanent tooth size of
Am 2001;45:253‑70. four human ethnic groups. Arch Oral Biol 2009;54 Suppl 1:S79‑85.